INSTITUTE OF PHYSICS PUBLISHING CLASSICAL AND QUANTUM GRAVITY

Class. Quantum Grav. 20 (2003) 93–99 PII: S0264-9381(03)39719-9

Cosmic acceleration with a positive cosmological

constant
Arbab I Arbab1
Department of Physics, Teacher’s College, Riyadh 11491, PO Box 4341, Kingdom of
Saudi Arabia

E-mail: arbab@ictp.trieste.it

Received 19 July 2002

Published 12 December 2002
Online at stacks.iop.org/CQG/20/93

Abstract
We have considered a cosmological model with a phenomenological model
for the cosmological constant of the form
= β RR̈ where β is a constant.
For age an parameter consistent with observational data, the universe must be
accelerating in the presence of a positive cosmological constant. The minimum
age of the universe is H0−1 , where H0 is the present Hubble constant. The
cosmological constant is found to decrease as t −2 . Allowing the gravitational
constant to change with time leads to an ever-increasing gravitational constant
at the present epoch. In the presence of a viscous fluid this decay law for
is
equivalent to the one with
= 3αH 2 (α = const) provided α = 3(β−2) β
. The
inflationary solution obtained from this model is that of the de Sitter type.

PACS numbers: 98.80.−k, 98.80.Hw

1. Introduction

One of the puzzling problems in standard cosmology is the cosmological constant problem.
Observational data indicate that
∼ 10−55 cm−2 while the particle physics prediction for

is greater than this value by a factor of the order of 10120 . This discrepancy is known as the
cosmological constant problem. A point of view which allows
to vary in time is adopted
by several workers. The point is that during the evolution of the universe, the energy density
of the vacuum decays into particles thus leading to the decrease of the cosmological constant.
As a result one has creation of particles although the typical rate of creation is very small. The
entropy problem which exists in the Standard Model can be solved by the above mechanism.
One of the motivations for introducing the
term is to reconcile the age parameter and
density parameter of the universe with current observational data. Recent observations of
type 1a supernovae which indicate an accelerating universe, once more draw attention to the
1 On leave from Comboni College for Computer Science, PO Box 114, Khartoum, Sudan.

0264-9381/03/010093+07$30.00 © 2003 IOP Publishing Ltd Printed in the UK 93

94 A I Arbab

possible existence, at the present epoch, of a small positive cosmological constant (
). One
possible cause of the present acceleration could be the ever-increasing gravitational (constant)
forces. As a consequence, a flat universe has to speed up so that gravitational attraction should
not win over expansion. Or alternatively, the newly created particles give up their kinetic
energy to push the expansion further away.
The purpose of this work is to study the phenomenological decay law for
that is
proportional to the deceleration parameter. In an attempt to modify the general theory of
relativity, Al-Rawaf and Taha [17] related the cosmological constant to the Ricci scalar, R.
This is written as a built-in cosmological constant, i.e.,
∝ R. A comparison with our ansatz
above for
yields a similar behaviour for a flat universe. And since the Ricci scalar contains
a term of the form RR̈ , one adopts this variation for
. We parametrized this as
= β RR̈ ,
where β is a constant. The cosmological consequences of this decay law are very attractive.
The law finds little attention among cosmologists but it provides a reasonable solution to the
cosmological puzzles presently known. We have found that a resolution to these problems
is possible with a positive cosmological constant (
> 0). This requires the deceleration
parameter to be negative (q < 0). Usually, people invoke some kind of a scalar field that has
an equation of state of the form p < 0 where p is the pressure of the scalar field. A more
recent review for the case of a positive cosmological constant is found in [6].
A variable gravitational constant G can also be incorporated into a simple framework in
which
varies as well, while retaining the usual energy conservation law [1, 10, 11]. The
above decay law leads to a power-law variation for G. Inflationary solutions are also possible
with this mechanism, thus solving the standard model problems. We have recently shown
that a certain variation of G may be consistent with palaeontological as well as geophysical
data [15].

2. The model

For the Friedmann–Robertson–Walker metric, Einstein’s field equations with the variable
cosmological constant and a source term given by a stress-energy tensor of a perfect fluid read

Ṙ 2 3k
3 + = 8πGρ +
, (1)
R2 R2
R̈ Ṙ 2 k
2 + + = −8πGp +
(2)
R R2 R2

where ρ is the fluid energy density and p its pressure. The equation of state is taken in the
form
p = (γ − 1)ρ (3)
where γ is a constant. From equations (1) and (2) one finds
d(ρR 3 ) dR 3 R 3 d

+p =− . (4)
dt dt 8πG dt
We propose a phenomenological decay law for
of the form [4, 5]
R̈

=β (5)
R
where β is a constant. Overdin and Cooperstock have pointed out that there is no fundamental
difference between the first and second derivatives of the scale factor that would preclude the
Cosmic acceleration with a positive cosmological constant 95

latter from acting as an independent variable if the former is acceptable [4]. Moreover, from
equations (1) and (4), one can write


8πG 3γ

R̈ = 1− ρR + R. (6)
3 2 3
Thus one example of
in the above form is the case when the universe is filled with a fluid
characterized by γ = 23 and β = 3. For other values of γ (=1), β is not constrained by the
Einstein equations, and the general relation


β

= 4πGρ (7)
β −3
shows the ratio of
to ρ is constant in this phenomenological model. Now equation (1)
together with equation (7) yield


β

= H 2. (8)
β −2
Thus, as remarked by Overdin and Cooperstock, the model with
∝ H 2 and the above form
(equation (5)) are basically equivalent. We see that in the radiation- and in matter-dominated
eras the vacuum contributes significantly to the total energy density of the universe in both eras.
Thus unless the vacuum always couples (somehow) to gravity, such a behaviour cannot be
guaranteed (and understood) at both epochs. Such a mechanism is exhibited in equation (20).
Hence the vacuum domination of the present universe is not accidental but a feature that is
present at all times. One would expect that there must have been a conspiracy between the two
components in such a way that the usual energy conservation holds. Therefore, one may argue
that in cosmology the energy conservation principle is not a priori principle [17]. We observe
 −29 
from equation (7) that

Pl0 ≈ ρρPl0 ≈ 10
1093 = 10−122 , where ‘0’ refers to the present and ‘Pl’
refers to Planck era of the quantity, respectively. Thus such a phenomenological model for

could provide a natural answer (interpretation) to the puzzling question why the cosmological
constant is so small tody, rather than just attributing it to the oldness of our present universe.
For the matter-dominated universe, γ = 1 and therefore equations (2), (3) and (5) yield
(for k = 0)
(β − 2)R̈R = Ṙ 2 , (9)
which can be integrated to give


A(β − 3) (β−2)/(β−3)
R(t) = t , β = 3, β = 2, (10)
(β − 2)
where A = constant. It follows from equation (5) that
β(β − 2) 1

(t) = , β = 3. (11)
(β − 3)2 t 2
Using equations (1), (5) and (10), the energy density can be written as,
(β − 2) 1
ρ(t) = , β = 3 (12)
(β − 3) 4πGt 2
and the vacuum energy density (ρv ) is given by

β(β − 2) 1
ρv (t) = = , β = 3. (13)
8πG (β − 3)2 8πGt 2
The deceleration parameter (q) is defined as
R̈R 1
q=− = , β = 2. (14)
Ṙ 2 2 − β
96 A I Arbab

We see from equations (11) and (14) that for a positive
the deceleration parameter is
negative (for β > 2). For β < 2 the cosmological constant is negative,
< 0. It has been
recently found that the universe is probably accelerating at the present epoch. There are several
justifications for this acceleration. Some authors attribute this acceleration to the presence of
some scalar field (quintessence field) with a negative pressure filling the whole universe. And
this field has a considerable contribution to the total energy density of the present universe.
The density parameter of the universe (m ) is given by
ρ 2 (β − 3)
m = = , β = 2 (15)
ρc 3 (β − 2)
2
where ρc = 8π 3H
G is the critical energy density of the universe. We notice that the Standard
Model formula m = 2q is now replaced by m = 23 q + 23 . However, both models give q = 12
for a critical density. This relation has been found by several authors [1, 3].
The density parameter due to the vacuum contribution is defined as 
= 3H

2 . Employing

equation (8), this gives

β

= , β = 2. (16)
3(β − 2)
We shall define total as
total = m + 
. (17)
Hence equations (1), (7), (15) and (16) give total = 1. This setting is favoured by the
inflationary scenario. The present value of the age of the universe, deceleration parameter and
the cosmological constant are obtained from equations (10), (11) and (15)
(β − 2) −1 2 (β − 3) β
t0 = H , m0 = ,
0 = H 2, β = 2, β = 3.
(β − 3) 0 3 (β − 2) (β − 2) 0
(18)
(the subscript ‘0’ denotes the present value of the quantity and H is the Hubble constant).
Inasmuch as q < 0, m < 1 hence the low density of the universe is no longer a problem.
Moreover, in order to solve the age problem we require β > 3. Thus the constraint β > 3
may resolve both problems. Observational evidence, however, does not rule out the negative
deceleration parameter and the stringent limits on the present value of q0 are −1.25
q0
2
[8].
The case β = 2 represents an empty static universe with
= 0. For β = 0, the usual
expressions for FRW models are recovered. When β = 0, m0 = 1, which is inconsistent
with many observational tests on scales much too small to be affected by the cosmological
constant, e.g. dynamical tests for scales up to a few tens of Mpc. For β = 4 one obtains
t0 = 2H0−1 , m0 = 13 , 
2 10 −1
0 = 3 ; β = 12, t0 = 9 H0 , m0 = 5 , 0 = 5 . It is interesting
3
2

to note that when β → ∞, all parameters are finite, namely, R0 → t0 , t0 → H0−1 ,
→

H02 , m0 → 23 , 

0 = 3 . Thus the values of β which are consistent with
1

m0 = 0.3 ± 0.1

are
β = 4.0 ± 0.5
but the ages are very high. For instance, with a high, but still consistent with observational
constraints, value of the Hubble constant, H0 = 80 km/s/Mpc, and a rather generous
t0 = 15 ± 2 Gyr,
Cosmic acceleration with a positive cosmological constant 97

the values of β consistent with this are:

5.5 < β
19.
This suggests a best-fit value somewhere around β = 5, which would give the somewhat
high matter density of m0 = 0.44 and rather high age of t0 = 18.3 Gyr. We have recently
investigated the implications of a variable G on the Earth–Sun system and, contrary to what has
been believed, we have found that the palaeontological data are consistent with a variable G
provided that the age of the universe is t0 ∼ 11 × 109 years and that G ∝ t 1.3 [15]. However,
a recent value for the age of the universe from gravitational lensing suggests t0 ∼ 11 ×
109 years. Recent estimates from observations of galaxy clustering and their dynamics
indicate that the mean mass density is about one third of the critical value [9]. Thus if β = 0
then the present age of the universe can not be less than H0−1 . This constraint represents our
strongest prediction for the age of the universe.

3. A model with variable G

We now consider a model in which both G and
vary with time. Imposing the usual
energy-conservation law one obtains [1, 10, 11]
ρ̇ + 3γ Hρ = 0, (19)
and
˙ + 8π Ġρ = 0.

(20)
Using equations (10) and (11), equations (19) and (20) yield
ρ(t) = Dt −3(β−2)/(β−3) , D = const (21)
and

(β − 2)
G(t) = t β/(β−3) , β = 3. (22)
4πA(β − 3)
 2/3
For β = 0, G = const and ρ = Dt −2 and R = 32 At , which is the usual FRW result.
Clearly for β > 3 the gravitational constant increases with time while for β < 3 it decreases
with time. Once again the constraint β > 3 considered before implies that the gravitational
constant increases with time. An increasing gravitational constant is considered by several
workers [1, 10, 11, 14].
In a recent work, we showed that the variation of the gravitational constant is consistent
with palaeontological data [15]. The gravitational constant might have had a very different
value from the present one. This depends strongly on the value of β assumed at a given
epoch. The development of the large-scale anisotropy is given by the ratio of the shear σ to
 
the volume expansion θ = 3 RṘ which evolves as [16]
σ
∝ t (3−2β)/(β−3) , (23)
θ
and, since β > 3, this anisotropy decreases as the universe expands and this explains the
present observed isotropy of the universe.
For example, if β = 32 then G ∝ t −1 , R ∝ t 1/3 , ρ ∝ t −1 . This behaviour of G was
considered by Dirac in his large number hypothesis (LNH) model [12]. In an earlier work, we
showed that some non-viscous models are equivalent to bulk viscous ones. This behaviour is
also manifested in our present model provided one takes β = 3(2n−1)
(3n−2) , where the bulk viscosity
(η) is defined as η = const ρ n , where 0
n
1 [1]. The two models, though different
in the form of
, are equivalent if one puts α = 3(β−2)
β
. Thus the decay law
= β RR̈ and

= 3αH 2 , where α = const, are identical in the presence of a cosmic fluid.
98 A I Arbab

4. An inflationary solution

This solution is obtained from equations (1), (2), (3) and (5) with β = 3. One then gets
R R̈ = Ṙ 2 , which integrated to give
R = const exp(Ct) (24)
where H = C = const. Applying equation (24) to equations (5) and (20) and employing
equation (1), we get

= 3H 2 , ρ = 0, G = const. (25)
This is the familiar de Sitter inflationary solution (in the matter-dominated epoch). A similar
inflationary solution is obtained with β = 3 in the radiation-dominated epoch. Inflationary
models employ a scalar field (inflaton) to arrive at this solution. These solutions help
resolve several cosmological problems associated with the standard model (flatness, horizon,
monopole, etc) The inflationary solution resolves some of the outstanding issues of standard
cosmology.

5. Conclusion

In this paper we have considered the cosmological implications of a decay law for
that
is proportional to RR̈ . The model is found to be very interesting and apparently a lot of
problems can be solved. To solve the age parameter and the density parameter one requires
the cosmological constant to be positive or, equivalently, the deceleration parameter to be
negative. This implies an accelerating universe. However, the strongest support for an
accelerating universe comes from intermediate redshift results for type 1a supernovae. The
model predicts that the minimum age of the universe is H0−1 . The behaviour that
∝ t −2
is found by several authors. The gravitational constant is found to increase with time at the
present epoch. Our model predicts an inflationary phase in the matter-dominated epoch as well
as the radiation-dominated epoch. The cosmological tests for this model can be obtained from
those already investigated by us [7]. The choice between these models awaits the emergence
of the new data.

Acknowledgments

My ideas on this subject have benefited from discussions with a number of friends and
colleagues. I am grateful to all of them. I wish to thank the Omdurman Ahlia University for
financial support of this work. I would like to thank the anonymous referees for their critical
and enlightening suggestions.

References

[1] Arbab A I 1997 Gen. Rel. Grav. 29 61

[2] Beesham A 1993 Phys. Rev. D 48 3539
[3] Matyjasek J 1995 Phys. Rev. D 51 4154
[4] Overdin J M and Cooperstock F I 1998 Phys. Rev. D 58 043506
[5] Al-Rawaf A S 1998 Mod. Phys. Lett. A 13 429
[6] Sahni V and Starobinsky A 1999 Preprint astro-ph/9904398
[7] Arbab A I 1998 Astrophys. Space Sci. 259 371
[8] Klapdor H W and Grotz K 1986 Astrophys. J. 301 l39
[9] Peebles P J E 1986 Nature 321 27
Cosmic acceleration with a positive cosmological constant 99

[10] Abdel-Rahman A-M M 1990 Gen. Rel. Grav. 22 655

[11] Beesham A 1986 Nuovo Cimento B 96 17
[12] Dirac P A M 1937 Nature 139 323
[13] Kalligas D, Wesson P and Everitt C W 1992 Gen. Rel. Grav. 24 351
[14] Abdussattar A and Vishwakarma R G 1997 Class. Quantum Grav. 14 945
[15] Arbab A I 1998 Preprint physics/9811024
[16] Barrow J D 1978 Mon. Not. R. Aston. Soc. 184 677
[17] Al-Rawaf A S and Taha M O 1996 Gen. Rel. Grav. 28 935